The future of robots and software automation doing the task of humans is here . . . and it first blush it’s downright scary. Robots replacing factory workers (an old but accelerating trend); autonomously driven vehicles threatening to make truck drivers and taxi drivers obsolete; and machines reading thousands of legal briefs in minutes are just a few examples of scenarios that sound like a sci fi disaster from an employment line perspective.
Should we be afflicted with automation anxiety? Yes and no, according to Joe Davis, Vanguard's global chief economist, who detailed his findings on the subject during a nearly hour-long presentation on Wednesday at the Morningstar ETF conference in Chicago. Actually, his ultimate answer leans heavily toward the “no” camp while acknowledging that the current and coming disruption will create some losers.
To begin his argument, Davis laid out what he sees as three paradoxes defining the economy and the markets right now: low inflation but full employment; low growth but high valuations in the market; and low volatility but high uncertainty.
“Some of my colleagues in the profession view these as either unrelated or isolated events,” Davis said. “But they’re not. In my mind they’re touchstones of a much larger force that will fundamentally alter the lives of millions of people on this planet.”
Davis believes we’re not in a period of stagnation but rather one of technical disruption, with more disruption headed our way. “And the nature of that change will bring four new paradoxes in the next three years.” (More on that later.)
“I argue that economists and policy makers have to pivot from focusing on GDP growth as the sole benchmark of an economy’s performance,” he continued. “More isn’t necessarily better. By growth, I mean productivity and standards of living, output per person and household income.
“I believe that we’re at an inflection point where today we’re seeing a structural change in the nature of work that will greatly impact the labor market not just for the United States but for the global economy,” he said. “How work changes for all of us in the years ahead will be the trend that defines our lifetime.”
That’s heady stuff, but what does this all mean for you, me and people who’ve yet to enter the workforce (or this planet)?
Davis provided context by looking at the past and potential pending consequences of economic disruption. Techno-optimists, Davis said, believe job losses caused by automation will be offset by new job opportunities being created.
“Throughout history the techno-optimists have always been right, he noted. “It happened twice before in the United States in the move from the farm to the factory, and then from the factory to the office. They’d say this time is no different.”
On the other hand are the techno-pessimists who have a more troubling vision of a future where any job that’s routine and repetitive and predictable will be coded into a computer and automated away. They argue that millions of jobs will be lost within the next five years.
Davis cited an Oxford University study that concluded that based on technology currently used today, 47 percent of U.S. jobs will be eliminated due to automation. That’s 70 million American workers.
He further noted that similar studies based on methodologies developed by M.I.T. found even more sobering rates elsewhere: 69 percent of jobs lost to automation in India and 77 percent in China.
“Imagine a world where three-quarters of the Chinese economy is without a job,” Davis pondered aloud, and then added, “If this is the world we’re heading for, I don’t know who’s buying ETFs. Maybe the rich machines are. Needless to say, these are alarming prospects.”
The Plus Side
Intrigued by these scenarios, Davis and his team at Vanguard replicated the studies and crunched the numbers to examine the future of work and gauge whether we’re all doomed.
His basic conclusion? Human beings are remarkable at adapting, and the doomsday scenarios likely won’t occur. If anything, he said, automation will accentuate the value of the types of tasks—and thus, jobs—that individuals increasingly do, anyway.
“I think both camps—the techno optimists and pessimists—take too simplistic a view,” Davis explained. “Work—like life—is more complicated than that. My message is ultimately one of encouragement. Work has always been in an arms race between education and technology, and skills and technology, and this time won’t be different.
“But this time two things will be different: the pace of the change and the scope of the risk,” he continued. “We all need to change our mindset about the nature of technological change and how it will impact the global labor market. Most importantly, jobs don’t get automated away—tasks do.”
Davis believes those pessimistic studies predicting mass human employment displacement caused by machines and software suffer from three fundamental flaws. The first is the notion that technology only threatens and substitutes for human labor, but it never complements it. “That’s wrong; just think of your own jobs,” he said.
Second, the pessimists assume jobs are made up of just one critical task. But many jobs have more than 20.
And third, and perhaps most important, tasks within any occupation change over time.
Using data from the Bureau of Labor Statistics, Davis and his Vanguard colleagues found there are more than 18,000 daily work activities performed in the U.S. across 974 occupations. From there, they scrunched those occupations into 41 unique tasks and put them into three broad buckets—basic, repetitive and advanced.
The basic category includes things like growing, harvesting, digging, moving objects, and recording information.
The repetitive grouping comprises tasks such as inspecting, monitoring, assembling and processing information.
And the advanced bucket—the largest of the three broad buckets—entails training, developing teams, strategizing, thinking creatively, solving problems, and assisting/caring for others, among other things.
“The common distinctions between blue-collar and white-collar jobs lose meaning when you look at work this way,” Davis said.
In the U.S., only 10 percent of our collective time is spent on basic tasks, versus 80 percent during the period of the Civil War, according to Davis. By 1940, American workers did repetitive tasks 80 percent of the time in settings such as factories and in offices, which led to the rise of the middle class. Technology has slowly automated away those tasks ever since.
The advanced category comprises 26 of the 41 unique tasks delineated by Vanguard. “Rather than advanced, I like to use the phrase ‘uniquely human’ because human beings have a strategic advantage in doing these tasks regardless of how smart a computer ever becomes,” Davis said. “These tasks can’t be automated away by computers.
“Thinking creatively is the top job requirement for a number of occupations not even related to the STEM fields (science, technology, engineering and mathematics),” he continued. “Examples include cooks and chefs, sales managers, program directors, tool and die makers. Other jobs that are immune from automation can involve social skills. Arguably the most important skill in the future isn’t IQ, but EQ, or emotional intelligence involved in maintaining relationships, managing teams, and developing and coaching others.”
In that vein, he noted, 70 percent of occupations require people to think creatively.
Don’t Let The Wave Swamp You
Technology is moving so fast, Davis offered, that the big question is can people keep up w/ the accelerated pace of change?
“Every story I read assumes that human beings are standing still in the face of this change and it’s only a matter of time before the machines race ahead and that eventually we’re all automated away,” he said. “But based upon our analysis, that’s not what our data say. We found that in the past 15 years across every occupation in the country we’ve seen a 50 percent change on average in terms of the tasks done and our responsibilities at our jobs.”
To Davis’ reckoning, the change in working at more advanced tasks means more human capital and, consequently, a higher paycheck.
“When we look at occupations that way there are variations, but every single occupation has seen an increase in the amount of time spent in uniquely human work,” he said. “In my mind, this is evidence that human beings can race with machines.
“By our calculations, for the first time in human history people are spending 50 percent of their time doing uniquely human work,” he added. “I predict that at least within 10 years that percentage will be 80 percent. And that’s good news because that’s associated with higher standards of living.”
Sounds rosy, but of course some people will still get the shaft.
“We get there two ways, which incorporates some good news and bad news,” Davis explained. “Automation will reduce about 20 percent of the occupations that Vanguard looked at, and most of those jobs are outside of manufacturing.”
But Davis believes those job losses will be dwarfed by the demand for new jobs, particularly those focused on uniquely human tasks. And he posits that the three most important skills going forward will be creative intelligence, technology acumen and emotional intelligence.
“The demand for those jobs will be significant in the years ahead, and that’s how we’ll get to 80 percent [of tasks performed using so-called uniquely human tasks].
New Paradoxes
Davis’ conclusion is that the changing nature of work fueled by technological disruption will create four new economic paradoxes.
1) More automation, yet labor shortages
The amount of robots in the workforce will double in the next three to five years, Davis posited, yet the combination of the aging population and the strong demand for uniquely human work will mean there will be labor shortages in the U.S.
2) Labor shortages, yet low inflation
Offsets to wage inflation due to technology is lowering the cost of producing just about everything, Davis said, and the more we use technology the harder it is to generate 2 percent inflation.
3) Low inflation, yet high real interest rates.
Even though Davis said he had the least confidence in this prediction, he foresees we’ll see the same level of global economic growth and the same inflation rate in the next three years, and we’ll see higher real interest rates not because the Fed will raise rates in December and probably will be on hold for at least a year, but because productivity will slowly rise and that’s often associated with higher interest rates.
4) High real rates, yet lower short-term market returns.
“High real rates is good news for all long-term investors because near-zero rates have depressed expected returns in the efficient frontier across all asset classes,” Davis said. “But there is a catch: we’d be in for some rocky weather as we adjust to that period during the next two or three years, and I think there’s more risk in the equity market than the bond market.”
Davis’ presentation offered a thought-provoking look into one company’s view—in this case, Vanguard—on the potential outlook caused by the dizzying and, for some people, unsettling technological changes occurring in the workplace.
“The defining trend of our lifetime isn’t so much demographics or globalization, although those forces will continue, but it’s the changing nature of work,” he said. “I believe the future will present new paradoxes and more challenges, but it’s an exciting one because it’ll provide more opportunity.”
Let’s hope he’s right.